Summary Business Intelligence for Data Science (Master DSS)

Les 1 Introduction to Business Intelligence for Data
Science
• Definition of a Database: A database is a collection of data or
information, often accessed electronically.
• Purpose of Databases: Databases are used to support Online Transaction
Processing (OLTP), facilitating efficient management of various
transactions.
• Database Management Systems (DBMS): These systems store data
within databases and enable users and applications to interact with the
data effectively.
• Term Usage: The term "database" is used to refer not only to the data
collection but also to the Database Management System itself

Why we need Databases?
• Important for Apps: Almost all interactive apps need a database to store
data well.
• Many Uses: Databases are like building blocks for apps in lots of different
areas, not only in businesses.
• A Variety of information: All sorts of information, from medical records
and online store things to grades, sports scores, and even game data, can
go into databases.
• Practical: Databases are like handy tools that help keep information
organized and easy to find, which is important for many types of apps.




Modern Business and Data Warehouse Systems
Characteristics of Business conditions:
 Uncertainty
 Risk
 Concurrency – staying ahead of the competition
 Predicting customer needs
 Globalization

,The motivation for the creation of a Data Warehouse Systems
Central component in BI systems is the Data Warehouse (DW)
The role of DW systems in business support:
• providing quality business information:
• reducing business costs,
• monitoring and analyzing operations,
• increasing profits.
• support in management and decision-making:
1. Management Information Systems (MIS), 1980.
2. Decision Support Systems (DSS), 1990.

The main components from both concepts of MIS and DSS
systems:
1. OLTP (Online Transaction Processing) - refer to specialized databases
and systems that handle daily transactions. They encompass various
technologies, databases, and methodologies and are a foundational
component of many businesses’ information systems.
2. OLAP (Online Analytical Processing) - set of tools and techniques
used to perform complex analysis on data stored in Data Warehouse
or other structured repositories.
3. Data Warehouse - central repository that collects, integrates, and
stores data from various sources within organization. For data
transformation and structure purposes ETL processes are used.

* OLAP and Data Warehouse are closely related to DSS systems, while OLTP is more
related to MIS systems.

, Databases Vs. Data Warehouses




The concept of Data Warehouse Systems
Creators of Data Warehouse system concept:
• Bill Inmon (1992): “A Data Warehouse is a subject oriented1, integrated2,
time variant3 and nonvolatile4 collection of data in support of
management’s decision-making process.”
• Oracle Data Warehouse method: “A Data Warehouse is an enterprise
structured repository of subject oriented, time variant and historical data
used for information retrieval and decision support. The data warehouse
stores atomic and summary data.
1
subject oriented
Data are categorized and organized by
business topics, NOT by functional units as in
OLTP systems.

Example of possible Data Warehouse themes:
• sales (data about products, customers,
spatial and organizational structure, …)
• marketing (data about the market, products, customers, competition, …)
• production (data about products, customers, technologies, plans, …)
2
integrated
Data about a single entity is entered
and stored in one place.

A Data Warehouse represents a
centralized database:
• it contains data from all organizational units of the company,
• in a "standardized format", as data from operational databases is often
structured differently.
3
time variant
Data is organized through sets of ‘snapshot
states’.
Each snapshot pertains to a specific time
interval.